Hi,
Word completion seems to look for matching words on the document from
the cursor point and up, and then, from the cursor down. It's often the
case, however, that the matching word you need is right on the next
line, or even later in the same line, and it takes a lot of
ESC-pressing to get to it.
My suggestion is that you sort them by proximity (as it is now), but
look for words in both directions, so the words right in the next line
stand a better chance.
(I had the impression that back in 1.0 it worked that way. Did it?)
Hi,
With most tree view like controls on OS X, alt-click will recursively
unfold all levels enclosed.
Is something like that possible or planned with fold markers in TM?
Alt-clicking does not seem to to anything different than normal
click, folds inside the clicked fold stay folded.
Gerd
Hi,
Very new to TextMate, so please forgive if this is a dumb question or
already well-known. I saw there was a similar discussion back in
March, but not quite the same issue as this:
I have the following four source files at the same level in my codebase:
FUFlickrExport.h
FUFlickrExport.m
FUFlickrExportCover.h
FUFlickrExportCover.m
If I open FUFlickrExport.m from, say Xcode, and hit Cmd-Opt-Up Arrow
to go to the header, TextMate opens FUFlickrExportCover.h.
TextMate correctly opens the counterpart of FUFlickrExportCover.
{h,m}. It also correctly goes to FUFlickrExport.m from its header,
it's just going to the header of FUFlickrExport.m that's not right.
Is there a better place to report bugs than on this list?
Best,
Fraser
--
www.speirs.orgwww.connectedflow.com
Hi,
Two things that I find consistently not to my liking in how TextMate
handles Ruby code:
1) When you fold a (ruby) function it places a "---" marker at the end
of the non-folded first line to represent (presumably) the folded
section.
However, although the "---" isn't part of the logical line, if you
place the caret before the marker and hit Enter the marker is moved
down a line and inherits a folding arrow. This leaves the first line
of the folded function disjointed with a non-functioning fold arrow.
=> I'd like hitting Enter at the "---" to move the caret to a new line
below the folded function without affecting the folded section.
2) When you fold a function and use Ctrl+/ on the visible line with
the "---" marker to comment it out what actually happens is that t
comments out only the visible line. The rest of the function is not
commented if you expand it.
=> I'd like to it comment/uncomment the folded section as well.
I am using v1.1b17
Regards,
Matt
--
Matt Mower :: http://matt.blogs.it/
Are language definitions merged?
There are a few languages where I have some 'private' rules. Is it
possible to write them in such a way that these are merged with
whatever language definition is found 'higher up', so that I can
easily enjoy updates from the svn repository without having to modify
that definition each time?
Gerd
Has anyone experienced an increase in TextMate's performance after
upgrading to Tiger?
On Panther, for me, simple things like switching tabs take half a sec.
My specs:
Panther 10.3.9 on Powerbook 15"
1.5 GHz PowerPC G4
1GB DDR SD RAM
I don't know how to fix this but maybe someone else does: When I use
the "Show TODO list" command (ctrl+shift+T), german umlauts like ä,ö
or ü are getting displayed wrongly.
Cheers. Oliver
Hello.
I'm experimenting with crazy nested syntax coloring.
I can't seem to figure out how to make all the transparent background
colors interact properly.
Ideally I'd like to define a scope:
scope test = everything within "(...)"
Then set a syntax color of transparent background color.
scope: test = White 10% opaque
Then nest a bunch of them inside each other.
zero (one (two (three (four (five (six (seven (eight ))))))))
result: The transparent background color would compound and gradually
get stronger:
n/a (10% (20% (30% (40% (50% (60% (70% (80% ))))))))
If I could get this to work, I could color different block types with
a massive background color & the content colors would be based on the
block color.
EG: HTML tags = blue BG, embedded ASP = red BG, strings = yellow
BG ... string in html tags = green, string in asp = orange.
With what I know now I could simulate this by explicitly setting the
background colors for every possible combination of elements, but
that's a lot of work. And if I decide to change the hue of something
I have to manually recalculate all of those static simulations of
color interactions.
---
I can get things to interact with the background color.
If I change the BG color for the scope "source" all the transparent
things define their transparency based on that color.
Also the line highlight & selection colors.
---
Am I asking for too much?
Will this make performance suffer?
Is this already possible, but I just can't figure it out?
